In partnership with:
CME Group
Eurex

92 Clear and Explainable Trend Models with Andreas Clenow of ACIES Asset Management – 2of2

"Never accept the old-style, blackbox explanation - always understand what is going on." - Andreas Clenow (Tweet)

In our continued conversation with Andreas Clenow, we discuss the research and work that goes into making his trading model, how he deals with risk and drawdowns, and how he explains his models to potential investors. In this episode, we will dive into how clear and explainable a trading model should be, and what questions you should ask firms before investing in them.

Thanks for listening and please welcome back Andreas Clenow.

Subscribe on:

Subscribe in iTunes Subscribe on Stitcher Subscribe on Soundcloud

In This Episode, You'll Learn:

  • How Andreas does his research and what research he looks for in other firms

    "I very rarely invest in a manager I don’t know about beforehand." - Andreas Clenow (Tweet)

  • Why a track record matters and how it matters
  • How trend following has done over the last few years
  • How he builds his own trading strategies

    "My experience is that if you combine too many things in one product it doesn’t make for a good product to sell." - Andreas Clenow (Tweet)

  • What he does with position sizing and how important it is
  • How he allocates risk on his different models
  • What he does to measure the environment on trend following
  • What he thinks about sector allocation and diversification
  • "It’s also risky to look over firms that you are overly familiar with - you need to learn about multiple types of strategies and ideas." - Andreas Clenow (Tweet)

  • The kinds of risk management that he uses in his model
  • Where he uses stop losses and where he doesn’t
  • How he talks to investors about drawdowns

    "The most important thing is that the investors understand what they are investing in - they should be as knowledgable as possible." - Andreas Clenow (Tweet)

  • What questions investors should be asking him about his trading model
  • How to detect if a manager’s strategy has stopped working
  • Why people shouldn’t be scared of asking lots of questions

    "The strategy needs to be clear and explainable." - Andreas Clenow (Tweet)

  • What motivates him to work on his job
  • About the biggest failures of Andreas’ career

Resources & Links Mentioned in this Episode:

This episode was sponsored by Eurex Exchange:

Eurex-Logo.svg

Connect with ACIES Asset Management:

Visit the Website: www.ACIES-am.com

Call ACIES: +41 44 868 41 11

E-Mail ACIES: info [at] acies-am.com

Follow Andreas Clenow on Linkedin

"Most people read too many trading books and too few programming books." - Andreas Clenow (Tweet)

Full Transcript

The following is a full detailed transcript of this conversion. Click here to subscribe to our mailing list, and get full access to our library of downloadable eBook transcripts!

Andreas

…that is the key trick. How do you get that? 

Niels

Sure. No, that’s fine. Let’s talk more about the usual topics that I would discuss with the managers. It somewhat relates to how you then build a strong organization. Let’s assume that you overcome some of these financial challenges that you just mentioned. In your case, I wanted to ask you something more focused on research because that’s something that you’ve done a lot of.  

Investors, they will put a lot of emphasis on the research capability of a manager when making their selections. Clearly research is, to a large extent, the heart of a systematic manager. But you also have an allocator or investor hat to put on. So if we put that on for a little bit now, what kind of research and capability do you look for when you are looking to allocate money away? 

Andreas:  OK, limiting to systematic managers, because we allocate to a lot of different things. Something I shouldn’t say on a public podcast because that means the amount of calls and emails I’m getting is going to increase from… 

Niels:  Oh, most definitely, Andreas. 

Andreas:  Yes, don’t worry I have an assistant to forward them to. What do I look for? Well first of all, I very, very rarely invest in a manager I don’t know about before. That would take something exceptional. Even the ones I know, well, obviously I meet them at times first. Managers usually come to our office to present. That’s actually how I got to know a lot of the interesting people in the business – Nigol and other people like that. You look at not only what they present, their real results, of course, and you look at the personalities. Is this an individual you can trust with your money or not?  

Obviously, before you actually pull the trigger on making a large allocation you need to go down to a much deeper level. You need to go down and do all of your proper due diligence. If you’re an institutional investor like us, then you are expected to see a DDQ, for instance, which obviously a newer manager probably won’t have. Some institutional managers might not either. 

Maybe I should explain. A DDQ is a Due Diligence Questionnaire. It’s the standardized way of answering a lot of questions. It’s the hedge fund equivalent of an FAQ on the internet. It’s usually a large booklet. What is it, a hundred questions? It feels like it when I fill it in, anyway. 

Niels

Yes, there are a lot of questions there, for sure. 

Andreas

Which you have to verify, because anyone can write a document, right? In this document should be enough information that you can verify it, you can figure out the answers to the most important things, and you save a lot of time if someone has a DDQ already. 

Niels

So, am I hearing you right when you say that it’s more the individual… Usually a lot of firms are founded by one or two people and so is it more for you the key individual rather than the fact that they might have 50 PhDs on their staff? So I’m trying to gather from an institutional investor like you what is important. 

Andreas

The principle of principles are of very high importance because if you don’t like them then it doesn’t matter how good the rest of the thing is. You have to figure out who they are, can you trust them, how do they run their business? I find some, even among the ones I actually like, I find some really extreme divergences between managers and how they are as individuals, how they run the business. They have often very, very different views on things compared to each other, and yet I’m digging to find out what really fits in with what you want to do and who you want to work with. You can also very quickly sort out many people who you just don’t trust. You meet them once and you count your fingers when they leave the office. 

No names, but quite often I meet people whom I would never do business with again, or at all. Personal contact is important. If the fund seems to be too important or too self-important to deal with you properly then why would you want to deal with them? 

Niels

Yeah. 

Andreas

So you begin to see who are they, what they’ve accomplished, how long have they been in business, are the results real? Are they showing you, simulated results, are they showing you actual results, what’s the difference, what kinds of fees to they have, what kind of structure do they have? What’s the legal structure, are they an offshore fund in a dolled-up country where you can’t sue them if something goes wrong, or are they on shore in your local jurisdiction? You need to look at all of these kinds of details. 

Niels

Sure. Well let’s stay with your investor hat on for a little while and talk about something that you just mentioned which is performance and more specifically track records. Now-a-days there are firms that have 30, 40 years of track record. There are some people who have 5 years of track record. What’s important to you, as an investor, when it comes to the track record? Also, maybe keeping in mind that people who have been in business for a long time obviously will have made changes to the model anyway, and therefore it can be a little bit difficult to just look at that. Since you do a lot of testing, you do a lot of simulation, you run a lot of things, you might be in a good position to have some views on this. 

Andreas

That’s an interesting point there, because you say some shops have been around for 30 years or something, and they’re large organizations with a lot of people, a lot of staff. I think you got my point before, that I’m allergic to large organizations. That might not be the best thing.  

When organizations grow to a certain point, no matter how good they were to begin with, they tend to (not always), but they tend to become bureaucratic corporate machines. It doesn’t happen to everyone, but there’s always a risk, and that’s something to look into.  

Maybe they were started by a great entrepreneur, he did excellent things for 20 years, but you know what, maybe now he’s just on the board of directors and it is run by a whole bunch of directors who are not owners who run things differently. You have all the project managers and the bureaucrats stepping in. It’s happened. If you find it becomes close after a while, because of these kinds of problems - others are struggling because of this. I’m sure you know some of these shops that are around. So it doesn’t have to be a negative thing, but it can be. 

Of course if it’s the same people that have been running it for a very, very long time, maybe it’s a great thing, and maybe it means that they want to retire soon and want to leave it and then what? It’s just things to look into. Never assume that it’s great because it has been around for a long time or that it’s bad because it has been around for a long time.  

 

A five-year manager might be great. It could be an excellent manager that has been around for five years. Of course you have to verify how it has been managed for four or five years and if this is mostly a retail account with their own money, I would be more skeptical. It doesn’t mean I would have to dismiss it, but that would be a larger hurdle of course. If this person has been running a proper audited hedge fund, with a proper custodian, and proper administrator and all of these things for five years, well that’s a decent track record. Five years running a proper business is not bad.  

 

Niels:  Yeah, I agree. Just speaking about performance but in a trend following context I just wanted to hear your view on that since you obviously, I can say I guess, have a soft spot for trend following, even though you have other strategies. A question I get from listeners from time to time, and actually also from potential investors that I deal with in my day job, and that is that people think that trend following, perhaps has performed poorly since the financial crisis because too much money has gone into that strategy since the great year of 2008. As a practitioner, what do you think about that? 

Andreas

Well, since 2008 there’s been good years and bad years. Overall, most funds made very good money since then. Most funds beat equities; they beat most normal classic types of strategies since then. It’s a favorite pastime of journalists, at times, to write about when trend following stops working again. All it takes is a few months, even a year that they make some great headlines, it’s always come back. Certainly we don’t have the same performance as in the 80s or even in the 90s. I don’t think it’s reasonable to expect that, but what we’ve seen is it still outperforms traditional asset classes. 

Niels

Yeah, I agree with that. Now, I actually never asked people about this, but since you’re wearing a number of different hats I wanted to ask you: we as managers often think about various statistics and how to measure how good a model is, or how good a program is. I’m wondering, and I’m sure you do the same when you do your own strategies, but also as an investor, I’m sure investors have their favorite statistics that they look at. What have you found in the world of many, many different statistics that seems to be meaningful and reliable to look at when evaluating a strategy or a manager? 

Andreas

That depends very much on what we’re looking for. I see the difference is, the perception is that we’re always looking for the best thing, but you’re never really looking for the best thing, you’re looking for something that fits what you need at the moment. Take an example: maybe you come to me and you want to pitch a new fund that has brilliant returns for a long time, real money, you have this enormous sharp ratio, and all of these things. You combine these ten strategies that have negative correlation to each other and therefore you have this great track record all of the time. That may not be what I’m looking for.  

As a Chief Investment Officer I’m probably looking for a building block that fits into my other building blocks. Someone comes and says I have the perfect thing because I combine all of these things. Well, you know, combining these things is my job, not your job, if I now have the Chief Investment Officer hat on.  

You find the same thing with most institutional allocators. They’re not looking for the investment strategy to rule all investment strategies. They’re looking for building blocks, they’re buying a Lego there. So the question is what do you need at the moment? 

Well probably you’re looking for something that has a certain profile in comparison to your current portfolio. Therefore, your validating both products and new models (if you build them yourself) on the merits of how do they compare to what you have at the moment? How do they fit in? Probably you’re not looking at replacing everything you have with this new thing, right? This is a portfolio component. If you’re the asset manager, you’re the fund manager and you have all these positions behind your fund, right? Well, when you come to me and you pitch your fund, then your entire portfolio is just a portfolio component for me. It’s just one component of many. So I think that’s the difference. I and I think most others in this situation, we don’t look at who’s got the best sharp ratio, or who made the most money last year. That is a simplified view. That’s not how it works. 

Niels

Let’s jump to the next subject, which is more the trading strategies. Now I need to ask you to take your investor hat off and put on your trader hat, or whatever we call it. Now, obviously, as people will have realized now, you run a number of different strategies: classical managed futures, some stock only strategies; so I want to ask you about how you (and you can pick one, or talk about both of them, it doesn’t really matter) generally speaking, how have you organized these strategies and what are the key design features that you have focused on? As you know, building these things is not strictly like following a recipe of a cookbook. There are some common ingredients, but you still have to put in your own choices, so how have you done it in your own trading? 

Andreas

Well, when you say, “my own trading,” that opens up for… 

Niels

Interpretation. 

Andreas

That opens the question, do you mean trading my own money, or trading client money, or trading accounts that are combined? 

Niels

I’m talking about the strategies that you would allow external money to be involved in. 

Andreas

OK, yeah, good. My point is once you’re talking about mandates where external clients are involved, either there is one single client behind it, and then that clients wishes and needs are the major factor. You never just say, “Take it or leave it. This is my model,” right? You discuss with a client, if now this is a single client, a single account. Given the client has many different things, maybe he has a large amount with us. We do many different things that can vary between investments in the American fracking business to, as you know, we’re heavily into merchant financing for instance combining with systematic strategies. So it depends on what is he looking for and what kind of risk and what kind of profile.  

Niels

I’m probably thinking about your systematic strategies, because that’s really been the theme of our conversation. 

Andreas

No, I get you. I just want to say that you have a marketing factor in this as well. Strangely, if you have the other type, which would be a collective scheme of some sort - a founder of a collective investment, in that case, of course, the marketing attractiveness of the strategy has to be a factor. Obviously if I do whatever I want with my own money it might look different. It doesn’t make it better or worse, but you have to take into account. How do you combine it or how do you decide? 

Well, if it’s something that external investors can come into, especially really outside money which might be on loan from people even, then it needs to be clear and explainable. No one wants to buy a black box system. You don’t say look at my brilliant returns, now just put the money in and don’t ask questions. You can’t run a business like that anymore.  

So you to be need clear and explainable. You need to have concepts that anyone can get. Personally I prefer to explain everything. If someone is interested, and frankly very few people are, but if someone comes and says, well maybe they are going to invest a million and maybe not, but can you explain the rules? Well, sure, come sit by my desk and I'll show you what we do, why not? As opposed to, for instance, if you have an account with just in-house money. Maybe you combine many different types of models, as I said, in ways that makes sense mathematically, but my experience, it doesn’t make for a good product to sell, if you combine too many things in one product or one account. It doesn’t become explained well enough. 

Niels

Sure, sure. Would you say, generally, taking this purer approach, does that mean that you… So today I guess that you do run a systematic trend following strategy, and you do run a systematic momentum for stocks? So in terms of that… I know your books talk about entry signals and how to aggregate signal generation and so on, and so forth. I guess there are three things that often comes to mind when you think about what makes a trend following system successful, if I use that as an example. It’s obviously where you enter, it’s where you exit, but it’s also how you manage the risk along the way. That lends itself to the position sizing as being quite important.  

You can talk about the signal generation, if you want, but I was interested in learning more about what you have found, for your own programs, to be important when it comes to position sizing and also whether you think that this is actually quite important in the determination of the success. 

Andreas

I would say that with position sizing you need to have a model that makes sense, first of all. You mentioned my slightly provocative post, blog post not long ago. I think you had one there a while ago about... it was just an easy target. I was just writing some things about so-called modern management strategies that you find so many books about.  

Niels

I haven’t read that one, so go ahead and indulge us. 

Andreas

Essentially it’s all this pyramiding and whatever else. It’s models on how you change your position size based on recent performance that seemed to be the basis of that series. For me that doesn’t make any sense at all.  

Let’s say if you, if you buy a position today, you have great gains in a week or two, then because you’re now “playing with the banks money”, right? So now you increase your position. You double your position size because you made a lot of money. It doesn’t make any sense. Why would you add higher risk just because you had a good run? The last week’s performance probably doesn’t impact the probabilities going forward to the next week, so why would you have so much higher risk on?  

I use very different models. Actually, I would say what I use is quite common models in the business, but quite different from what you read in most retail trading books. I prefer to have some variation of the volatility parity slicing. You look at some sort of volatility measurement. Which one, for this discussion, is not very important, but you can use something as simple as ATR, if you like. You allocate a certain risk, if you want to use volatility as a proxy for risk of course, but it’s close enough, and you take a position based on that.  

The only problem is, of course, if volatility is not static neither is your portfolio value. So, if you’re in long-term positions, you need certain intervals, or certain threshold levels, rebalance your positions and reset them to whatever risk you would like. It doesn’t mean that you buy when you make money or you sell when you make money or vice-versa, but that you’re targeting a risk level, if that changes, for whatever reason (your current risk), you need to keep track of what risk you want to have here and adjust your position size back to that. 

Of course that’s a position level, but, as I like to stress as well, the most important thing is at the portfolio level. What you mentioned here with entry signals and exit signals and the position size, this is all position level. I would say, for the systematic models, diversified systematic models in a way that trade many markets and many instruments, every single position is quite irrelevant in the end.  

It’s not the result of an individual position that matters, only the end result on the portfolio level. So that’s also a mistake that I see quite often is overly focused and developed trading models. They focus on the position level, but the only thing that matters for you at the end of the year is how much gains, or what happened on your whole portfolio. Who cares if you lost half of your positions or even more if you make money and it’s good risk adjusted values on the whole portfolio. Risk management and risk allocation on the portfolio level is where you should focus. 

Niels

Is that very different between the two types of models that you’ve written about now in your two books, The Classical Trend Following Model and the Momentum Model for Equities is the portfolio level and the way you allocate risk on that very different? 

Andreas

You know what, yes. When you trade equities you have to be aware that your primarily trading Beta. That is the dependency on the index is what’s important. So you will always have very high dependency on what’s going on in the overall market and index. Therefore, your risk looks different. The long equity model will always, in a way, be a relative model. You can’t expect to have great gains when the market is falling. It doesn’t mean that it is either worse or better, or that you make less money or more money, it’s just different. You can’t expect the same type of returns so you can make money in any type of market, which you potentially can with futures. You won’t, but potentially you can. 

Niels

Sure. So clearly position management and position sizing is important. Just from your experience, exits – where people get out, where does that rank in your world as to how successful a strategy is? 

Andreas

I would tend to say not very for the reason that we’re still dealing on a position level. As an example, let me take a super simple example. I mentioned before this type of counter-trend model I made for a firm where you buy, where trend following models exit. I made a funny variation on that in the beginning mostly as a joke, it simply exited after 30 days, no matter what – nothing else. Who cares what happens between 30 days, take profits, we close our eyes until then. That works great. So you have to look at a model like that and look at that model and you tell me the exit is more important than the entry, then I’m not sure about that. It’s hard to say anything absolute - which is the most important, overall for all trading models. Well, there’s no such thing. It all depends. 

The boring answer to everything is, “It all depends.” 

Niels

What about the environment, meaning if you could measure the environment, because I think you use something you refer to as trend filters in your recent book and so I imagine that you could have something along those lines, at least I have it on my website, I have a trend barometer that measures the environment for trend following strategies and obviously people are welcome to look at that, but do you find that some kind of filtering, or some kind of measure of the environment to be part of a strategy now-a-days? Is that required, or not necessary? 

Andreas

It’s certainly helpful, particularly for equities, to have something like that. You have to be aware of where the wind is blowing. It doesn’t mean that you have to necessarily take a position in that direction, but if you have some sort of trend filter, you have some sort of idea where things are more likely to move on a certain time horizon.  

It’s the same thing whether you look at the market as a whole or whether you look at each market by itself. You have to be aware, when you trade natural gas, you have to be aware of what’s there. What’s the long, medium, and short-term timeframe countertrend on this at the moment. It doesn’t mean you have to trade the same direction, but you have to be aware of it.  

Niels

In terms of diversification we’ve talked about that for the classical trend following, you feel that it’s very important for the strategy that you have the underlying diversification. Where do you stand on the overall asset allocation among sectors, meaning do you need to allocate the same risk to every single market, every single sector, can you be overweight in certain sectors? How do you deal with that in your own systematic trend following portfolio? 

Andreas

With a risk of starting with, “it depends” again, it depends on what type of model. Say you’re looking at an equity model, for instance. It can, to some degree, ignore the sectors. It doesn’t matter what type of model you have. If your model has the potential, say, of going into 100% health care stocks, for instance, you probably want to prevent that.  

For some types of models, maybe you want to have some sort of sector restraints. I would say it’s a bit too far to look at tracking error and that type of approach. I’m not a big fan of the mutual fund way of doing things. Some sort of sanity filter, let’s say if you apply for instance the racking model I use in the latest book, and if you apply that on, say, instead of S&P 500 you apply it to NSCI wall stocks. Maybe you come up with that you should now have 85% of your stocks in Japan, well then you have to wonder, do you really want to do that? Maybe you do, maybe you don’t but it’s something to consider if you really want to allow those things to happen.  

For the futures side, I think it’s less of a concern. A futures model often depends on deliberately taking corner risks. More often than not you make the really big returns on corner risk. 

Niels

How do you define corner risk? I’m not familiar with that term. 

Andreas

Corner risk, basically is high-risk concentration in a particular area. So let’s say, what if, for instance most things are not trending, but the crude oil is dropping heavily and you’re short to crude futures, so you’re short the gas/oil, the gasoline. You’re short the heating oil. You’re short the whole complex messily correlated, but you make money on all sides for a while. Suddenly you take a hit on all of them and you exit. This is a necessary evil of diversified trend following, I believe. 

There are probably some smart people out there that have figured out ways around it. Even if you have to take the corner risk like that, my experience is that you’ll make money from it in the end. 

Niels

Sure. Now I want to go to the next topic, which is risk management. Before I do so I just have a couple of questions. One is, we’ve talked a lot about momentum trading, counter trend and trend following. You may even do more strategies like that. I have just one simple question, if you could only choose one to trade, which one would you feel most comfortable trading? It’s like choosing between your children, isn’t it? 

Andreas

Well, I only have one. 

Niels

So that’s easy. 

Andreas

Yeah, true. I was about to give you the boring answer. The most profitable things we’re doing is actually outside of systematic strategies. Anyhow, you want to hear about systematic strategies, so therefore, let me say normal trend following on futures probably has the best potential. If I were to pick only one type, I would say go very long term, but it’s very slow trend following, which is, over the long run, usually the best one. You’ve seen, as well, some of more famous managers that have been around a long time running very, very long-term trend models. They make them seem like heroes some years, and lunatics other years. In the end, despite people occasionally seeing them as lunatics, they tend to come out as winners after enough time. 

Niels

Sure. Now, of course, now that you bring that up, it’s sort of reminds me that in your latest book about trading equities, you actually suggest (and I know the days completely irrelevant) but you actually suggest to trade every Wednesday. 

Andreas

Oh yes, I had a good reason for that. It’s a 20% probability of being the right day, right? 

Niels

Absolutely. What you’re saying is, I just want to make sure I understand it correctly. Are you saying you should run your systems every day, but using very long term parameter sets, or you should actually just run a trend following model once a week and not worry too much what happens in-between? 

Andreas

I have both kinds, actually. Look at something like my interpretation of this 12-month momentum model, it checks a signal once a week. Not weekly data, because why not use high frequency while you have it, right? It just only trades once a week, and then checks the value a year ago and yesterday, and that’s it. So that, overall, performs very well and there are unpractical problems with running it, it’s a very theoretical model.  

As for the book, by the way, I mention that as well. I could be good to mention that to a wider audience. When you write a book you always have some sort of regret afterwards and realize I should have done something different, right? As I said, the first one I should have simplified the model more. The second book, well I should have, in retrospect, decreased the trading frequency a bit.  

Living where I live, you know what I’m saying, countries, we don’t have to deal with capital gains taxes, even for private trading. As investment professionals we certainly don’t have to deal with it. We’re dealing with very low commissions and so on. Therefore, trading frequency is not a big deal. For the book, given that a lot of people who read the book are located in countries where they have capital gains taxes to deal with, even this so called short term capital gains taxes in some countries, the amount of high commissions and so on, if you instead of trading weekly were to trade monthly, you’d get almost the same performance and much lower trading costs, much less issues with capital gains taxes. So yeah, in retrospect, I should have probably done monthly in the book instead of weekly, but I should do a post on that. 

Niels

Sure. Now jumping to the risk management side, how do you define risk? What is valid risk to look at, in your opinion, when you are running these strategies? 

Andreas

It depends on how deep you want to go. 

Niels

I think you like that term, Andreas, “It all depends.” 

Andreas

It all depends, it does. Reality tends to be more complex than the theories. Besides, I’m always careful when talking about risk in public, since I have a wife who is a professional risk analyst, risk controller, risk manager, and so on. I should be careful using the right terminology. How far should we go? It depends on who you are and what kind of situation you are in.  

Niels

Let’s talk about how you do it, not your private money, but when you manage money for clients, what kinds of risk meshes do you use - standard deviation, valued risk, margin to equity, stuff like that? 

Andreas

Exactly, yes, all of the above, of course. The question is then how do you use this in the end? I know some firms, for instance that use valued risk to allocated positions, and they look at marginal of value risk of new positions to the performance, and so on. I don’t really. That gets too complicated and I haven’t seen, for me, the benefit. I can see the logic and I can certainly see how that fits some special larger organizations. Also it gets these kinds of things… running these kinds of models gets much more important if you have a large staff.  

Obviously if you have a portfolio where you have one or two primary managers where every single position they look at every day, the need then for expensive risk systems, a need to have an MCI borrow system or SunGard system, it decreases a lot because they understand the risk anyway. You can build your own things, even in Excel if you have just one or two or even five portfolios and a small staff. Once you have 20 people or more looking at it, you need some sort of way of aggregating risk on a company level and we need to follow-up on a whole different type of methodology. 

Niels

Are you a believer in stop losses? Meaning that every single trade should have a stop loss, or do you manage the risk of loss in a different way yourself? 

Andreas

Generally I’m against the concept of every single system should have anything… For some type of trading approaches, I use stop losses, for others I don’t. For equity markets, for instance, I tend not to have stop losses in the traditional way. I have other ways to make sure when a position should be halted or out and replaced with something else or closed down, but not the traditional type of stop loss. For some futures models I find stop loss to be absolutely vital, not all but in some models, especially those kinds of counter trend models. I find a higher importance of a stop loss since they are actually going in and buying in the middle of selling pressure, for instance, and if this is actually something real, it could be a big move.  

So different models would require a different version of that. 

Niels

Yes, it all depends, doesn’t it? 

Andreas

I guess it does. 

Niels

Now, drawdowns, Andreas, have you found anything to set the expectations with investors in terms of drawdowns? When you look at a certain strategy, and again I’m referring to maybe your momentum stock trading, or it could be your traditional trend following trading. Is there anything as a guideline that you found is a good measure or good estimate of drawdowns? This is obviously one of the things that investors get very nervous about that is when you start losing money. 

Andreas

The most important thing is that investors understand what they invest in. They have to understand what kinds of drawdowns are possible, what kinds of drawdowns are likely, and of course, whatever drawdowns are simulated, your real ones are going to be larger and most likely are going to be coming around sooner than you think. That’s just how things work, right? So the most important thing in order to retain investors is to have them as knowledgeable as possible about your strategy, about the risks, about the drawdowns. If you hope that you won’t get a drawdown in a year or two, and you don’t properly explain the risks to a client, well, when that drawdown comes, he’s not going to stay. Communication, as usual, is the main point. 

Niels

Sure. Just going back to the original question, is there anything from doing a lot of simulations, etc. etc.? Do you just go with whatever your 20-year simulation says and then add another 50%, or have you found anything else that works in terms of estimating the actual potential drawdown, or do you not worry about that as long as you explain to investors that this is to be expected and these are the environments that we would expect these losses to occur? 

Andreas

First you have to look at what’s likely. For that you can use simulations, you can look at the worst thing that happened before, assume something a little bit worse will happen and that’s probably likely to happen. You see recovery periods and how it handled those things. Always assume that it’s going to do worse than whatever your worst simulation is, but then of course, you have the scenario of something absolutely unexpected happening and it gets much worse than that. You have to have some sort of plan in place for what happens when the unexpected happens. 

What happens when you show up to the office one day and I don’t know what, a nuclear bomb went off, something huge happened and everything is just way off the charts, now what? No one wants to think about it. We all have to think about it. You have to have some sort of plan for it. 2008 is a good example as well. Things happened that no one had planned for.  

Niels

Is that, in your opinion, where systematic trading really has its strength. There is always a plan. Every day there is a plan as opposed to maybe a discretionary trader. Is that really the crux of the matter? 

Andreas

For a normal bad market, yes. Take this week for instance. This is great for systematic traders. Win or lose you know what to do, right? It was, from a market point of view a bad week, but not a panic week of any sort.  On the other hand, systematic traders without special plans, have no answer for what they are going to do in say in the autumn of 2008. Yeah, you can look back and say they did great, right? But classic models, many of them broke down in various ways, especially the risk management got insane in many of them that they didn't mess with the override.  

But there's even bigger problems coming up that no one ever thought about before. Like what to do with all the cash. Certainly there's a game of hot potato. No bank is safe. That was a perception in a way for a while. There was a couple of months, I think around four or five months where I spent every late afternoon figuring out which bank is the safest one for the next 24 hours and then we moved money that we really needed to have cash on hand on fund accounts, client accounts, moved them within 24 hour deposits between different banks and that was a stressful period. 

Obviously if you just left the cash in your account and run your trend following model and sit back and watch, well maybe you show up one day and your brilliant strategy has blown up because your bank went bust and you lost all of your cash. Guess what, 75% of your fund was in cash and now you're dead, right? These kinds of things, you have to have... If not before '08 it's kind of forgivable if people didn't have plans for these things before 2008 because it was unimaginable before, but now we know that anything can happen and plan for it.  

Niels

Is this the kind of risk that would keep you awake today. I don't mean necessarily that you think any bank will go bust tomorrow, but I'm just curious, you systematized a lot of your trading so is there anything that you worry about when you go to sleep every night when it comes to risks in your trading? Something that you accept that you can't program? 

Andreas

If you worry when you go to sleep at night then you have a problem. You're not going to last long in the business I think. I think you have to get over those things. Obviously you have to deal with all the problems, but when you can't let go of the problems once you're away, then there is a problem, then there is a concern that you're not going to mentally last in the game if you go to bed worried over risk every evening then that's not healthy, that's not healthy. Find a way to deal with it during work hours. The glory of spending 80 hours a week in the office and being there evenings and weekends, that's an illusion you cannot get over once you get to a certain age. Been there and done that and glad I don't have to do that again. 

Niels

Sure, sure. Now I've got a couple of topics left for our conversation that I just want to run by you. I want to talk briefly about research. Again, you have these two hats to wear. So this time I'm going to ask you to wear your allocator and investor hat again. When you sit down and you have your research brainstorming sessions with your colleagues and you're trying to decide what to invest in, be it the manager strategy, etc. etc. The way we make progress is often to ask good questions. What are the questions that you think investors should be asking when it comes to the overall asset allocation of their portfolios? 

Andreas

That's a broad question.  

Niels

How do you do it? It could be how you do it when you look at your portfolio of liquid strategies. I know you do some other stuff, but of the liquid stuff where it involves hedge funds, CTAs, what are some of the questions that you ask yourself? 

Andreas

The first thing is to look at what you want to accomplish here? What's the aim? No, making money is not the aim. It's too simple. Usually either you're looking at some low risk capital preservation strategies in the long run, or you're looking for capital building wealth, building account strategy.  

So you have to decide what you want to do. Do you want to have a high-risk portfolio? Do you want to have something long term? Are you trying to structure something for a family office that intends to stay here five generations from now? How much risk do you really want to take in order to get some money? What kinds of building blocks do you need? How do they fit together? How much of each type of building block might you need? Maybe you need some high-risk blocks, some low risk blocks, combine different strategies.  

It's always a risk to look at too much at things that you are overly familiar with. If you're really into trend following then you have to analyze 10, 15, 20 trend following shops and you invest in the best ones. You’re still just in trend following, in that case. Maybe you want to be in trend following but maybe you shouldn't only have that, right? 

You need to learn about multiple types of strategies, multiple ideas and be open to new types of ideas. In my experience there are usually new areas, new types of strategies both liquid and illiquid that come up from time to time, that might, at first, sound a bit strange, somethings we may not know about, but maybe this is actually a rising opportunity for a while. So keep an open mind and look at a lot of different options. 

Niels

Sure. I know that was a very broad question, so forgive me about that. One question that is maybe not so broad, but also quite relevant, and that is, when you as an investor look at systematic strategies and probably look at the ones that you have invested in. How do you detect, (because this is a question that I get from the listeners often) how do you detect if a manager's strategy has stopped working? It goes back to the point about, is trend following dead? There's obviously going to be times where things look like they've stopped working but they may not be. How do you really determine when it's stopped working and I don't mean trend following as a concept, I just mean a particular strategy? 

Andreas

Yeah. If the returns are not what you think they're supposed to be, you need to have a chat with the manager. You need to get an explanation and see if it's a good one or not. Maybe the manger is changing something, maybe there's something unexpected happening, maybe there's just something in the market that you weren't really aware of how it would impact – take something like the near zero yield environment that has created quite a few problems for the futures space in general. Maybe there's some factor that you're not aware of and you need to get a good explanation from the manager.  

If you are really convinced that the results are off and should not be like this. Everyone has a loss, but maybe there's a loss in a period where you think there shouldn't be a loss, and you ask the manager about it and he doesn't have a good explanation for it. Maybe you should be careful. Maybe he's right, maybe he's not, but if you're not comfortable with it then exit, do something else. Always understand what is going on and never accept the old style black box explanation. Don't worry, just sit down and trust me and give me a year and I'll fix it – you want real explanations.  

Niels

Sure. Talking about explanations and questions and so on and so forth, you've been part of many due diligence meetings, I'm sure, and calls in your time, and both people looking at you, perhaps, and you looking at other people. Is there anything that you've found that investors really should be asking the managers they look at, and perhaps they're not really asking often enough? Is there anything in particular you found which is really important to get a good debate on as part of looking at a manager? 

Andreas

I find too often that people are a bit scared of asking too many questions. There's always this perception that the guy in front of me has so many secrets and it would be impolite to ask about them, but there's no harm in asking. Besides, if you don't get an answer you better get a good reason why you don't get an answer.  

Say for instance, ask for something like, what shows on trade charts. I want to see some positions where you enter and exit and why. It's all these kinds of things that you can get explanations for. If they say we combine these two strategies or two time horizons, or something, OK what's the attribution of each? How about the attribution per market, per long short, per sector? You can get into a lot of these details and you'd be surprised how many of the really big, the really great famous asset managers, and many of them that you've had on these calls would give you an answer, it's just that most people don't ask them. Most good managers will give you good answers to those kinds of questions. They will invite you to come and sit by the screen and show you everything. OK not the source code, but other things. 

Niels

Sure. 

Let's jump to the final topics of our conversation today, which is more the general and fun stuff, now that we've been so busy talking about systems and statistics and all of that. You have been doing this for a while. It's probably more a job to you than anything else. Why do you keep doing what you do? What motivates you? 

Andreas

It's fun. I like doing what I'm doing. There's got to be a reason to do things. It's like they used to say, "Life’s too short to be stuck in a crappy job all day." Most of your life you spend working one way or another. Do you really want to be sitting in a cubicle somewhere yelling at people in teleconferences? They might give you a nice title, you feel important with the company for a while, but is it something that you want to do for 20 years, or 30, or 40, or I don't know what, then fine. If you like it, then fine. But having fun, liking what you do is important.  

I do a lot of different things from writing books, I manage money quite a lot of different ways, I have my website where I regularly upset people, of course. I publish various analytics. I automate our futures analytics report from the website and these kinds of things and I have fun doing it. I can, I have the freedom to do things and I do what I think is fun. That's important. 

Niels

Sure, sure. Now, if you are going to recommend a book other than the ones that you have written. 

Andreas

Why would I do that? 

Niels

Exactly... To people who want to improve their trading, so let's stay with the finance, the trading theme here, what books have made a big impact on you. It may be one. It may be a couple. Are there any ones that spring to mind? 

Andreas

Starting with what made an impact on me, they have to go back in time to Sweden, to Gothenburg again. I read a lot of trading books back then, but most of all I think Jack Schwager’s earlier books had a big impact on me back then.  

Niels

Sure. The Market Wizard books you mean? 

Andreas

Exactly, the Market Wizard books, I like other books too, but the Market Wizard books, the two first ones in particular they had a big impact on me back in the mid-90s I guess, which is the reason I pursued him... maybe you saw the back of my first book, there's a statement from Jack. I was very happy to manage to track him down and get him to do that afterwards. He was one reason why I got into the business in the first place. 

Another book just to mention, it's not necessarily because it's a great book, but it has had quite an impact on me. In an airport in London, I believe, some 15 years ago I found a book called something like Create and Manage a Hedge Fund, something along that line, which I bought. I think that opened my eyes to begin with to the fact that, you know what, I can start a hedge fund. That was a great buy. It was one of those big, thick, heavy books. 

Recent books, I like the quite recent book by Katy Kaminski, called Trend Following... 

Niels

Trend Following With Managed Futures.  

Andreas

Yes. Thank you. Sorry Katy, I should have known that. 

Niels

That is a great book and of course Katy has been a guest on the show here, so by all means should go back and listen to that episode. 

Andreas

Absolutely. Let me see, another recent book, Gary Antonacci, wrote a good book this year about momentum strategies as well. It's quite different from mine but it's a very good book and very different from mine so I think they are good compliments because they come from very different perspectives.  

Niels

Who was that who wrote that? I didn't get that? 

Andreas

Gary Antonacci, it is called Dual Momentum 

What else? I've got some on my bookshelf, but it's too far to read the titles. I guess I have to... 

Niels

That is fine. You gave a couple of good suggestions. If you have more, you can always email them and I'll make sure that I put them on the show notes page for this. 

Andreas

There's one thing, generally speaking, just generally speaking you find more value in reading a programming book than a trading book, generally speaking. There's some great trading books and there's some bad programming books, but most people read too many trading books and too few programming books. 

Niels

OK. I've never read a programming book, myself, I have to admit, so maybe I should do that. 

Andreas

I'll give you some pointers. 

Niels

Now what's been your biggest failure in your career, Andreas? 

Andreas

Should I take my time and write you a list? 

Niels

Exactly. You thought we were just about to finish now, and I come up with this kind of question. 

Andreas

That's a tough one. I don't know. Everyone has had failures. Everyone has had trading losses. For instance, I was fully loaded in Japanese stocks during the tsunami. That was not a fun day. Japan was the hottest stock market in the world and suddenly we had a tsunami and nuclear accident and those were not fun losses on that. Most of these things seem like the end of the world when they happen, but give it a year, you recover, you come back, and it doesn't seem as big anymore. You always have these incidents where they seem like the end of the world. 

Niels

Yeah, that's very true. Now, you alluded to a secret of yours earlier on, but I'm going to ask you for another one and that is, is there a fun fact that you can share about yourself? Something that even people (and I don't mean your wife) but people who know you may not know about you. 

Andreas

You have to give me some preparation for these kinds of questions. 

Niels

I know, then it would be scripted. It would be boring wouldn't it, so take your time (at least three seconds). It could be a talent. Do you have a hidden talent that... maybe you sing like Pavarotti when you're in having a shower, I don’t know? 

Andreas

At the risk of continuing a stereotype, I did once walk into a chess tournament as the only known club player and walked home with a prize. Of course it was a small town and not that many players, but it still annoyed everyone that a club player, the guy who was able to fill out the list of people, walked home with the prize. I always used to be a pretty good computer game player, but who wasn't a long time ago? I had time for that. 

Niels

So a secret hidden talent of Chess. 

Andreas

I'm not sure how secret any of those things are, but yeah. Do what you want with it. 

Niels

Now you mentioned that you have a son and so if you could pass on just one of your own skills to your son, what would that be and why? 

Andreas

Probably not the computer game skill. Oh, come on, this is too difficult a question. 

Niels

I know. As soon as it gets personal, Andreas, it starts getting a little tough here, doesn't it? 

Andreas

Oh yeah, of course. 

Niels

What has served you well in your career? 

Andreas

I'm guessing my inability to conform. I'm the classic non-conformist. I don't like... you tell me there's a certain set of environment and rules and things to follow, and again, I'm not talking about the actual ones - laws and these things, but rather if you put me in a corporate environment and draw a circle or box around my responsibility areas, well, if I feel it makes sense to do something outside of there I will. If it makes sense to leave that box completely, I will do so. It’s going to be very difficult to keep me there.  

People like that, people like me are often very difficult to manager if you’re a manager. Of course that doesn't apply to any future employers who I might offer me a job after I've lost all my money. I believe, to be serious, I think it's an important skill not to accept nonsense in life, but rather make up your own rules, make up your own reality and refuse to accept whatever is pushed upon you. 

Niels

Sure. When your son grows a little bit older I'm sure you will play this segment for him when he makes up his own rules at home and you don't agree with them then at least he has evidence on his side. "Daddy said I should do this," right? 

Andreas

I guess so. 

Niels

I said earlier today, when we spoke, that it's important to ask a lot of questions, and the right questions, and so on and so forth. So I'm going to turn it a little bit on myself at the very end here and just ask you whether there's anything that you think we've missed today in our conversation, something you want to bring up from your books, from your trading, just to make sure that I do justice to you and your firm. 

Andreas

I think you pretty much covered it. 

Niels

Good, that was easy. Now as we wrap up, I want to thank Eurex, the exchange, for sponsoring today's episode. As many of our listeners will know, Eurex is a great place to go if you’re a systematic trader because there's lots of liquidity around and especially in times of volatility, like we see at the moment, it's certainly one of the favorite exchanges for many systematic traders. Before we finish our conversation completely, Andreas, can you tell me about where people should go if they want to learn more about your books, or your activities as such, where should we send them? 

Andreas

Sure. Just go to my webpage at FollowingTheTrend.com and you can contact me there. You can read my articles. You can click on my links and buy my books. I'll be there. 

Niels

Very straightforward. From my side, as a little bit of housekeeping I just want to mention before we finish this episode that you might, as a listener, want to just check your subscription button on your smart phone, because we don't publish as regularly as we did a little while ago. Sometimes these media players will unsubscribe the podcast, so I encourage you to just go and check that you are still subscribed to the podcast so that you automatically will get all the future episodes. 

That leaves me to say, thank you very much, Andreas. It's been a great conversation, a bit of a master class in systematic trading, which I really appreciate and your willingness to share all of your insights, and being so open about specifics and rules. If people will go and buy your book, which I encourage them to do, they will find some great starting points into this world. The latest book is Stocks on the Move, by Andreas Clenow. As you grab that copy, I would also suggest you buy the Classical Trend Following one that Andreas mentioned. All of the details from this episode you, the listener, can find on the show notes page on Top Traders Unplugged. 

So Andreas, I know we will connect at a later stage either on a podcast or locally here, in Switzerland, so I wish you a great evening and I appreciate your time. 

Andreas

Thank you, thank you very much. 

Niels

All the best. 

Ending

Thanks for listening to Top Traders Unplugged. If you feel you learned something of value from today's episode, the best way to stay updated is to go on over to iTunes and subscribe to the show so that you'll be sure to get all the new episodes as they're released. We have some amazing guests lined up for you, and to ensure our show continues to grow, please leave us an honest rating and review on iTunes. It only takes a minute, and it's the best way to show us you love the podcast. We'll see you next time on Top Traders Unplugged.